Development of an Immunochromatographic Test to Detect Antibodies in Simulated Whole-Blood Samples for the Diagnosis of Intestinal Capillariasis

ABSTRACT.

Intestinal capillariasis is a zoonotic parasitic disease caused by Capillaria philippinensis, a fish-borne nematode. Patients present with clinical symptoms that include borborygmus chronic diarrhea, intermittent abdominal pain, weight loss, and several degrees of painless lower-leg edema. Atypical clinical manifestations and diagnostic confusion with diarrhea caused by gastrointestinal cancer, HIV infection, giardiasis, and hyperthyroidism may occur. Misdiagnosis and improper treatment can be fatal. Parasite eggs and larvae are passed intermittently in feces, rendering fecal examination insensitive. Serological methods can serve as a supportive diagnostic tool. In this study, we developed a simple and rapid antibody detection lateral-flow immunochromatographic test (ICT) as a point-of-care diagnostic tool. A new ICT format using a Trichinella spiralis larval extract antigen for antibody detection in simulated whole-blood samples (WBSs) was evaluated, and the results were compared with those obtained using serum samples. The sensitivity, specificity, and accuracy of IgG antibody detection in simulated WBSs were 100%, 94.3%, and 95.7%, whereas in serum samples, these values were 100%, 91.5%, and 93.5%, respectively. The results between simulated WBSs and serum samples did not differ significantly, with a concordance of 97.8%, yielding a Cohen’s kappa of 0.95 (P <0.001). The ICT is easy to implement and can provide supportive diagnosis at the bedside or in local and remote hospitals with limited facilities.

INTRODUCTION

Intestinal capillariasis is a zoonotic parasitic disease caused by Capillaria philippinensis (C. philippinensis), a fish-borne nematode.^1,2 Human infection has been reported in many countries across Asia³ and Europe,⁴ with an increase in reported cases in recent years.^5–8 The Philippines and Thailand are endemic areas for intestinal capillariasis,^6–9 with sporadic reports from Egypt, India, Iran, Japan, Korea, Taiwan, Italy, Spain, the United Arab Emirates, and the United Kingdom.^5,6,9 An imported case from an endemic area has been reported in Europe.⁴ Fish acquire the parasites from the fecal matter of infected birds or from environmental contamination by infected humans. Humans can become infected by consuming raw or undercooked small freshwater fish. Once ingested, the worms can multiply within the body because capillariasis uniquely allows autoinfection, leading to severe hyperinfection. Patients with intestinal capillariasis exhibit symptoms such as borborygmi (rumbling sounds caused by gas moving through the gastrointestinal tract), flatulence, chronic diarrhea, and intermittent abdominal pain.^3,5 Complications may include protein-losing enteropathy which can lead to peripheral edema, malabsorption, electrolyte imbalances, and weight loss.^3,5 Misdiagnosis and inadequate treatment can occasionally result in death due to electrolyte imbalance or septicemia from secondary bacterial infections.³

Intestinal capillariasis poses significant challenges in clinical practice.² Diagnosis is difficult because of atypical clinical symptoms and the potential for confusion with other causes of diarrhea, such as gastrointestinal cancer, HIV infection, giardiasis, and hyperthyroidism. Additionally, patients may be unrecognized by clinicians if the illness occurs in a nonendemic area. Standard fecal examination exhibits low sensitivity, as parasite eggs and larvae are purged irregularly.³ Definitive diagnosis is performed by investigating the characteristic parasite eggs, larvae, and adult worms in stool samples.³ However, Capillaria eggs can be mistaken for look-alike Trichuris trichiura eggs by inexperienced personnel, and nematode larvae in fecal samples are difficult to identify as C. philippinensis.³ Endoscopy and imaging assist clinicians in diagnosis by enabling the direct visualization of parasites or facilitating intestinal biopsies; however, these methods are costly and invasive.^2,10 Serodiagnostic tools, including antigen detection using capture ELISA,¹¹ antibody detection via immunoblotting,¹² and ELISA¹³ using the Trichinella spiralis larval extract, can support diagnosis because this antigen shares antigenicity with C. philippinensis. The latter two antibody detection methods showed 100% sensitivity.^12,13 However, immunoblotting and ELISA require sophisticated equipment and are labor-intensive, limiting their use in rural areas. Recently, an antibody detection tool, the rapid immunochromatographic test (ICT) kit has been developed for the diagnosis of intestinal capillariasis in human sera.¹⁴ However, this ICT tool was optimized for use with serum samples and had not been evaluated with whole-blood samples (WBSs). To address this gap, we evaluated a new ICT designed to detect antibodies in simulated WBSs as a rapid and reliable diagnostic tool for intestinal capillariasis. Diluted blood samples were applied directly to the ICT strip with no requirement for a washing step. The diagnostic performance was evaluated under laboratory conditions and compared with the values obtained using the corresponding serum samples.

MATERIALS AND METHODS

Clinical samples.

The conduct of this experiment was guided by the criteria for Standards for Reporting of Diagnostic Accuracy Studies 2015.¹⁵ Leftover serum samples stored at −70°C from the frozen biobank at the Department of Parasitology, Faculty of Medicine, and the Mekong Health Science Research Institute Biobank Project at Khon Kaen University were used for this study. A total of 231 serum samples were divided into three groups and manipulated to simulate WBSs. Group 1 consisted of 55 samples from patients with intestinal capillariasis from the northeast region of Thailand, which is a nonendemic area for trichinosis. These patients were parasitologically confirmed to be infected with C. philippinensis through stool examinations conducted using the formalin-ethyl acetate concentration method,¹⁶ which was used to detect eggs, larvae, and adult worms of the parasite. No other active infections were shown. All individuals in this group reported a history of consuming undercooked fish and experienced symptoms that included intermittent diarrhea, abdominal pain, and a gurgling stomach for several weeks. Importantly, none of the patients exhibited trichinosis-related symptoms, such as periorbital edema, fever, or myalgia, suggesting that these cases were likely not associated with trichinosis. Group 2 consisted of 40 serum samples from adult volunteers who acted as control subjects. Fecal examinations were performed in duplicate using the formalin-ethyl acetate concentration method,¹⁶ and no intestinal parasite infections were identified when blood samples were collected. Group 3, the other parasitosis group (n = 136), consisted of serum samples from patients with conditions other than intestinal capillariasis (Table 1).^16–22 All infected individuals received standard treatments according to the criteria established by the Ministry of Public Health of Thailand. Simulated WBSs were prepared by spiking the serum samples with group O red blood cells (RBCs) according to the described methodology.²³ The RBCs were prepared from blood group O, which was donated by one of our researchers, in an ethylenediaminetetraacetic acid anticoagulant. The 3.5 µL of packed RBCs was mixed with 6.5 µL of each serum sample to simulate WBS with the normal levels of human blood components. Additionally, pooled positive and negative control WBSs were prepared by combining equal volumes of sera from 10 intestinal capillariasis patients and 10 healthy volunteers, respectively, with packed RBCs, which were added in the same manner as outlined above.

Table 1.

Results of the immunochromatographic tests for anti-Capillaria IgG antibody detection in simulated whole-blood and serum samples

Categories of Samples		Diagnostic Tools	Number of Positive Cases/Total Number Examined (range of band intensity levels)
			Simulated Whole-Blood Samples	Serum Samples
Capillariasis group	Intestinal capillariasis	FECT16	55/55 (0.5–7)	55/55 (0.5–8)
Healthy controls group	Healthy Thai individual	FECT16	0/40 (0)	0/40 (0)
Other parasitosis group	Strongyloidiasis	FECT16	0/10 (0)	0/10 (0)
	Angiostrongyliasis with eosinophilic meningoencephalitis	Serology17	0/10 (0)	0/10 (0)
	Gnathostomiasis	Serology18	0/10 (0)	0/10 (0)
	Ascariasis	FECT16	1/10 (1)	2/10 (0.5–3)
	Hookworm infection	FECT16	0/10 (0)	0/10 (0)
	Trichuriasis	FECT16	2/12 (0.5–3)	4/12 (0.5–3)
	Trichinosis	Serology19	7/10 (0.5–5)	8/10 (1–5)
	Giardiasis	FECT16	0/10 (0)	0/10 (0)
	Amoebiasis	FECT16 and serology20	0/10 (0)	0/10 (0)
	Blastocystosis	FECT16	0/10 (0)	0/10 (0)
	Cysticercosis (Taenia solium)	Serology,21 CT, and MRI	0/10 (0)	0/10 (0)
	Opisthorchiasis	FECT16	0/10 (0)	0/10 (0)
	Taeniasis (Taenia saginata)	FECT16	0/10 (0)	0/10 (0)
	Sparganosis	PCR22 and histopathology	0/4 (0)	1/4 (0.5)
Diagnosis values	Sensitivity (%)	–	100.0 (95% CI [93.5–100.0])	100.0 (95% CI [93.5–100.0])
	Specificity (%)	–	94.3 (95% CI [89.8–97.2])	91.5 (95% CI [86.3–95.2])
	Positive predictive value (%)	–	84.6 (95% CI [73.5–92.4])	78.6 (95% CI [67.1–87.5])
	Negative predictive value (%)	–	100.0 (95% CI [97.8–100.0])	100.0 (95% CI [97.7–100.0])
	Positive likelihood ratio	–	17.6 (95% CI [9.6–32.1])	11.7 (95% CI [7.2–19.0])
	Negative likelihood ratio	–	0.0	0.0
	Accuracy (%)	–	95.7 (95% CI [92.2–97.9])	93.5 (95% CI [89.5–96.3])
	ROC area	–	0.972 (95% CI [0.954–0.989])	0.957 (95% CI [0.937–0.978])

CT = computed tomography; FECT = formalin-ethyl acetate concentration technique; MRI = magnetic resonance imaging; PCR = polymerase chain reaction; ROC area = receiver operating characteristic area.

Rapid diagnostic test kit.

Muscular larvae of Trichinella spiralis (T. spiralis; ISS62)²⁴ were extracted and prepared as previously described¹² and used as the antigen. The developed kit for intestinal capillariasis, which detects specific IgG antibodies in either simulated WBSs or sera, was designed as previously reported.²³ The kit consisted of five key components: 1) a nitrocellulose membrane (on which 0.4 µg protein/line of T. spiralis somatic larval extract was dispersed as the test line and 0.4 µg protein/line of goat anti-mouse IgG [Lampire Biological Laboratories, Pipersville, PA] was dispersed as the control line using a XYZ3210 Dispenser [Bio-Dot, Irvine, CA]), 2) a conjugated pad (colloidal gold-conjugated mouse monoclonal anti-human IgG was sprayed onto a glass microfiber filter [Whatman Schleicher & Schuell, Dassel, Germany] to form the conjugated pad), 3) a sample pad, 4) an absorption pad, and 5) a backing card. All the components were laminated and cut into strips 4 mm wide using a guillotine cutter (CM5000 Guillotine, Bio-Dot), and each strip was then enclosed in a plastic housing (Adtec Inc., Oita, Tokyo, Japan; Figure 1). The completed ICT kit was stored in a desiccated, sealed aluminum foil package at room temperature until use and demonstrated stable reproducibility and consistency for up to 24 months when stored at ambient temperatures ranging from 4°C to 30°C. The complete kit consisted of an ICT strip and chromatography buffer.

Figure 1.

Immunochromatographic test for intestinal capillariasis diagnosis. The result is read at 15 minutes using the naked eye, and color intensity is assessed using the interpretation card (left; a test line intensity level of ≥0.5 indicates a positive result). The presence of both colored bands on the test and control lines indicates a positive result (middle). If a colored band appears only at the control line, it indicates a negative result (right).

To use the ICT kit, each simulated WBS or serum sample was diluted (1:10) with the provided chromatography buffer. Subsequently, 5 µL of diluted simulated WBS or serum sample was added to the sample well, and 60 µL of chromatography buffer was added to the buffer well. All results were recorded 15 minutes after the start of the assay. The within-day and between-day precision of the ICT kit were determined via the analysis of pooled positive and negative reference samples. A positive result was indicated by the emergence of a red band at the test line at 15 minutes (Figure 1). Additionally, a red band consistently appeared at the control line to confirm the functionality of the kit. The intensity of any band was assessed visually without assistance using an interpretation card, with the threshold level set at ≥0.5 to indicate positivity (Figure 1).

STATISTICAL ANALYSES

Diagnostic values were calculated according to the previously described methodology.²⁵ All statistical analyses were performed using Stata Statistical Software: Release 10 (StataCorp LP, College Station, TX). The software was used to determine the key performance metrics of the diagnostic kit, as well as the sensitivity, specificity, positive and negative predictive values, positive and negative likelihood ratios, accuracy, receiver operating characteristic areas, and Cohen’s kappa value. The Cohen’s kappa values were interpreted according to established guidelines: scores of <0.20 indicate poor agreement, scores of 0.21–0.40 represent fair to near medium agreement, scores of 0.41–0.60 denote moderate agreement, scores of 0.61–0.80 signify good agreement, and scores of 0.81–1.00 indicate almost perfect agreement.

RESULTS

The ICT kit for antibody detection demonstrated high performance in both simulated WBSs and corresponding human serum samples, as detailed in Table 1 and Supplemental Table 1 and Supplemental Figure 1. In simulated WBSs, the kit achieved a sensitivity of 100%, a specificity of 94.3%, and an accuracy of 95.7%. Similarly, when tested with serum samples, the kit exhibited a sensitivity of 100%, a specificity of 91.5%, and an accuracy of 93.5%. The agreement between the results obtained from simulated WBSs and serum samples was assessed using Cohen’s kappa value, yielding a value of 0.95 (P <0.001), which indicates almost perfect agreement. Additionally, the concordance rate between the two sets of results was 97.8% (226/231).

DISCUSSION

In this study, a newly developed ICT tool for the detection of anti-Capillaria IgG antibodies in simulated WBSs and serum samples is reported for the rapid diagnosis of patients with intestinal capillariasis. The ICT strip used T. spiralis larval extract as the antigen because these antigens are also shared with C. philippinensis.^12–14 Similarly to C. philippinensis, T. spiralis is classified within the superfamily of Trichuroidea, the members of which possess a stichosome that produces similar cross-reacting antigens.^26–29 Additionally, the current ICT demonstrated a diagnostic sensitivity of 100% for antibody detection in both WBSs and serum samples. This aligns with our previous reports, which also revealed 100% sensitivity for serum-based detection.¹⁴ This high sensitivity is likely because the intestinal capillariasis patients evaluated in this study were chronic cases, as the disease typically presents clinical symptoms and signs during the chronic stage. Consequently, the samples were possibly collected during this period, resulting in a strong antibody response. Future studies should be conducted to evaluate the ICT’s performance in early infection cases to assess its effectiveness across different stages of the disease.

The absence of any false positives in the WBSs and serum samples from healthy individuals provided high diagnostic specificities, ranging from 91.5% to 94.3%. However, the specificity of this ICT test was slightly lower than previously reported (96.6%).¹⁴ Various factors, such as types of buffers, membranes, cover housing, antigen concentrations, and differences in the sample panels used, may contribute to variations in diagnostic values between tests. Some simulated WBSs cross-reacted with cases of ascariasis (1/10), trichuriasis (2/12), and trichinosis (7/10), whereas corresponding serum samples cross-reacted with cases of ascariasis (2/10), trichuriasis (4/12), trichinosis (8/10), and sparganosis (1/4). Although cross-reactions with ascariasis, trichuriasis, and sparganosis sera were observed, this is unlikely to pose a problem in a clinical setting because these helminthiases typically present with clinical features that are different than those of intestinal capillariasis. The positive reaction for trichinosis might seem concerning; however, trichinosis presents with facial edema and myalgia, symptoms that are not seen in capillariasis, which reduces the likelihood of misinterpreting the ICT results.³⁰ Moreover, the cross-reactions of the ICT kit might lead to an apparent false positive result in sera from other helminthiases because of asymptomatic intestinal capillariasis cases,⁵ because of serum samples collected from intestinal capillariasis patients in whom coproscopy failed to detect the parasite,³¹ or because patients had previously undergone incomplete anthelminthic treatment, resulting in no detectable parasites.

However, limitations should be noted. First, the current ICT test kit has only been developed and evaluated under laboratory conditions using a defined set of frozen samples in a single-site study. Its performance may vary depending on the types of diseases that could potentially cross-react in the regions where the kit is deployed. Therefore, the kit still needs to be evaluated using a larger number of samples from other parasitic infections and real blood samples because diagnostic values might differ according to the tested population. Second, the developed IgG antibody detection system is unable to discriminate between previous and active infections. Third, T. spiralis larval extract was used; thus, clinicians working in areas where trichinosis, trichuriasis, and ascariasis coexist should carefully interpret the diagnostic results. In these circumstances, when suspected intestinal capillariasis cases have positive serological test results, patients should undergo multiple stool examinations or receive anti-helminthic therapy to facilitate worm expulsion.³ The parasites usually exit the body through stool within 2 to 3 days,³ ensuring a confirmed diagnosis and reducing concerns about previous exposure to these helminths. Substituting antigens is beneficial only when the method is applied in regions where intestinal capillariasis is reported but trichinosis is not endemic. To improve the kit, new highly sensitive and specific Capillaria antigens produced through recombinant technology should be used. This approach would ensure a stable supply of antigens, simplify quality control, and likely enhance diagnostic values.

CONCLUSION

In conclusion, this ICT is optimized for use with simulated WBSs and should be applied to real whole blood, such as fingerstick blood obtained in the field via lancet puncture, in the future. The test exhibits high sensitivity and specificity, enabling diagnosis using both human serum and simulated WBSs. When utilizing simulated WBSs, there is no need for hemolysis or subsequent washing before applying the strip device. This test kit can support the clinical diagnosis of intestinal capillariasis at the bedside, facilitate diagnosis in poorly serviced remote areas, and serve as a surveillance tool in regions with a high risk of intestinal capillariasis.

Note: Supplemental materials appear at www.ajtmh.org.